d2r: why (not)echo is important -- part 2

why (not)echo is important -- part 2

Last week I posted a comment with my views on why, necho is important. It was done from a more technical point of view, but the arguments ended up in what's really important anyway: the users, because in the end it's about creating useful applications, interoperability, etc.

I was reading Steve Kirk's open letter to the RSS community and I thought I'd elaborate on one the points on that I have mentioned before, (and, as I've also noted earlier Jon Udell made similar comments recently), that I think is getting lost in the discussion, and thus muddling things a little. To see the rest of my argument, please refer to the previous entry I mentioned above.

Before going on: Steve puts forward the idea of creating a standards body for the current formats. I agree that would be the ideal case, but it's plainly clear that at this moment it can't happen. I think everyone, from their side of the fence, would agree with that as well. As I've said in one of the posts I referenced above:

I think there's no doubt that Echo's happening, which is good. Also, there's no doubt that, ideally, it would be better if the process involved less infighting and was more evolutionary.

So, the reality is that we were not going anywhere before. The reasons aren't important at this point, and I don't want to dwell into a discussion that has been going on for long enough, and one in which, plainly, there can be no "winners". What matters is where we are today, and how to go forward from here.

On to what I was really going after, what I consider a point that has been largely ignored in this "compatibility" discussion.

The point is this: creating a new weblog syndication/API format will not be disruptive for users.

Why do I say this?

Consider the state of things today: weblog tools generate feeds in any of several formats (sometimes in more than one). The BBC uses one format. The New York Times uses another. When tools or sites provide information in different formats, they are, in fact, incompatible.

Do users know this? Do they care?

No.

Why?

Because tool providers have evolved to deal with a situation of multiple formats, and support all of them transparently. I know, because I've written software that works that way. All other aggregators do the same.

So the fact is that, today, users are not being directly affected by the multiplicity of formats, because we, the developers, have evolved to support a splintered market in a consistent way.

In the previous paragraph I said "users are not being directly affected" because they are being affected indirectly. How? Mainly, through the extended development time that supporting multiple formats mean for developers, and consequently less time to be able to do new things.

So how does this reality affect the necho/RSS argument?

In my opinion, it gives us a good indication of what will happen when necho is "released". Tools will start to support necho as well as RSS. The formats will coexist, just as RSS 0.91 and RDF and RSS 2.0 coexist today. Furthermore, this coexistence will be transparent, just like today. Over time, necho will, hopefully, become the standard. In the meantime, there will not be a major catastrophe of incompatibility (although we can't rule out minor problems). Eventually, some of the other formats might become less used, and will be phased out (this is something that is already happening, for example, with the transition from RSS 0.91 to RSS 2.0). And because, currently, RSS is being almost exclusively used for updates and regenerated constantly at each endpoint, there will be little if any switchover cost, again, as an example of this I put forward the transition from RSS 0.91 to RSS 2.0 that happened last year. (This is a point on which I disagree with Steve, who makes comparisons to Linux and Windows, which I think is innacurate. The cost of switching binary formats is of a completely different order than the cost of switching RSS, as I've mentioned here, and as clearly shown by the RSS 0.91 to RSS 2.0 switch, which happened late last year).

Obviously, it's on us, the developer community, to add necho support without disruption, and it's not a problem. After all, we are already doing it today, and moving most (hopefully all) tools into necho will eventually reduce work for developers in the future, allowing us to, finally, concentrate on improving the tools rather than on how to let them connect to each other.

Note: As I said in the previous entry: this is an emotional subject for many people, so I'd appreciate it if the comments, if any :), remain on-topic, that is, they talk about the text itself, or the ideas, rather than about the people that stand for/against an idea, both for comments on comments, or comments on the entry. Thanks.

Categories: technology
Posted by diego on July 6 2003 at 5:47 PM

Comments (please see the comments & trackback policy).

Diego, I agree with some of what you said, but I think you missed the point on the Linux comparision. It had nothing to do with Windows, just the basic concept of using Linux in a business environment for production purposes. I did not write enough in my post to be clear. Linux use for things like file, print and other server-type purposes became more widely acceptable when RedHat got involved in a very public fashion.

As far as the rest of your post, consider the developer's burden. Very few companies make web browsers because of the large amount of standards to comply with (my opinion) and the fact that the product (web browser) has turned into a commodity. RSS readers, no matter the format that is delivered to them, will soon become a commodity. When the RSS reader becomes integrated with an email reader, Microsoft will jump in with a version of Outlook. That will mean that whatever formats Microsoft Outlook supports will become the "defacto" standard. It's not a Microsoft vs. the world issue as much as a standards issue. I don't care who sets the standard as much as caring about who has a seat at the table. An oversight group will fix that and make sure that everyone gets their "day in court".

Posted by: Steve Kirks at July 6, 2003 7:41 PM

Steve, thanks for the comment.

I see what you meant now with Linux, etc. However, I still think that the cost of switching RSS (if that was necessary) is hugely lower, and it can't be compared with changing OS platforms across organizations. Also, since end-users right now for the most part are not aware of the fact that different versions exist, there is even less of a problem.

Regarding "When the RSS reader becomes integrated with an email reader". Not to be self-promoting :), but you can check out clevercactus, the software I referred to in the post, which provides that functionality today (although the next version will provide WYSIWYG editing, but that's another story...). Back to the point: about MS, etc. I made a similar proposal to yours nearly two months ago on this weblog (In it, I mentioned that I was willing to sacrifice openness at the beginning of the process, to allow for stabilization, etc). That is: I agree with what you said. I don't care who sets the standard, just that one exists, and is accepted and embraced by everyone. Of course, if others, including me, can participate, all the better.

Posted by: Diego at July 6, 2003 7:53 PM

I very much liked the line : "creating a new weblog syndication/API format will not be disruptive for users". Quite. Absolutely no-one involved in the Necho project wants to cause problems for users because they *depend* on those users.

The developer's burden right now is ridiculous - supporting all those different dialects, using an approach to client-server comms (XML-RPC) that ignores (well, precedes) all the work that has been done in the past few years in web architecture and services. Having a single, consistent (syntax and API), unified approach to syndication is the best way we can go forward.

For those people that suggest otherwise, I think it might be worth examining what commercial interests they may have.

Posted by: Danny at July 6, 2003 8:19 PM

Tell me, what compelling user benefits does the 'clean new design' offer?

And Danny, your last comment is not helping this discussion in any positive way.

Posted by: Don Park at July 7, 2003 2:55 AM

Danny: "For those people that suggest otherwise, I think it might be worth examining what commercial interests they may have."

I suggest otherwise. REST is simply fashionable, unnecessary geekery, IMO. XML-RPC should be the basis of any unified API. It's what existing systems have in common, and it works.

Meanwhile I indeed have commercial interests. People pay me to provide them with a blogging/discussion environment.

Now tell me what one has to do with the other.

Posted by: Roger Benningfield at July 7, 2003 7:23 AM

Don: for what I think are benefits, please check the previous post I linked to on why I think necho is important. (Note, that the phrase you mention, "clean new design" has never been used by me in either post, of by the comments. I note this because "clean new design" implies that the old design wasn't clean, which isn't true at all, and I think everyone would agree on that. There were some elements that were a bit unclear--everyone would agree on that too.)

Additionally, as I've said in the post, given that the situation today is badly fragmented, and that users seem to be doing okay, I haven't yet seen a use-case in which necho would affect a user (or in which a user would even have to know that it exists).

Tools that read RSS/RDF today will read RSS/RDF tomorrow. Tools that post to metaweblog will also continue working.

And the fact that the change from RSS 0.91 to RSS 2.0 was done late last year without major problems proves it.

Roger, I think that Danny wasn't specifically relating XML-RPC or the API to the "commercial interests", but to the whole package. On the other hand, I think that there are commercial interests on all sides... anyway, I'd prefer not to start nitpicking like this... somehow it's easy for these kinds of arguments to get out of hand. :)

Posted by: Diego at July 7, 2003 10:37 AM

Diego: Don't worry, I don't care enough to get out of hand. :) There was just a note of "you're with us or against us" in Danny's comment that I didn't want to let pass. The folks working on Necho are generally good people trying to do a good thing. But one can disagree with their objectives or approach without sinister motives... good people can have bad ideas, after all.

Posted by: Roger Benningfield at July 7, 2003 2:36 PM